Google Gemini: Everything you need to know

0
[ad_1]

Google has an extremely powerful AI chatbot called Bard. It’s already proven to be a useful and very capable chatbot, and Google has integrated it into several of its products. While the company is working on Bard, it’s also developing Gemini. But, what is Gemini and how is it going to be an improvement over Bard?

That’s what this guide is going to go over. We’ll talk about what it is and answer any questions that people may have about it. This article will constantly be updated, so you should definitely check back every now and then to see what new capabilities have been added.

What is Google Gemini?

Think of Gemini as Bard on steroids. Better yet, think of it as Google’s equivalent to OpenAI’s GPT-4. GPT-4 is a much more powerful version of ChatGPT 3.5, and it will have the ability to generate more than just text.

This is similar to Gemini. Just like Bard, it will be able to generate text as per your input, but it will go further than that. In general, Gemini will be a set of powerful LLMs (large language models) that will work in tandem as an entity in and of itself. It’s going to be a much more powerful and capable version of Bard and probably more powerful than GPT-4.

Who is Gemini going to be targeted towards?

It seems that Gemini will be targeted towards business professionals who need a powerful AI tool. It’s expected to be more powerful than Bard, and it will be able to process a ton of data. We’re sure that average users will find a use for it, but its primary demographic could be business professionals.

Will Gemini be better than ChatGPT?

Right now, it seems that Gemini will be much more powerful than GPT-4. We will need to wait for both to launch in order to tell for sure, but Gemini could take the cake in this battle.

Gemini will be multimodal. What does that mean?

Multimodal means that a model is able to process and output more than one type of media. For example, a multimodal model will be able to output both text and images. This is expected to be the case with Gemini. It’s expected to be able to generate images based on your text or image input.

How many tokens will Gemini be able to process?

This information isn’t official, but rumor has it that Gemini could have up to 1 million tokens. Think of tokens as bits of information that a chatbot can “remember”. A token can be as small as a character and as large as a word.

If you type in “I had a bad day”, that’s four tokens, and the chatbot will remember that information when speaking to you. Say, if you type an entire novel into Gemini (let’s say the novel is 50,000 tokens), and Gemini can remember up to 100,000 tokens, then it will remember every bit of information in the book, and it will be able to use that information when generating its responses.

For context, GPT-4 can currently process up to 8,000 tokens. That’s more than enough for most queries, but it’s nowhere near what Gemini is rumored to do.

How many parameters will Gemini have?

Along with tokens, parameters are another aspect of an advanced AI model. This information is still rumored, but Gemini could have over a trillion parameters. GPT-4 is said to have up to 1.7 trillion parameters. We’ll have to wait until we see both models at their full potential.

Who has access to Gemini?

At the time of writing this article, Google is testing Gemini on a limited number of companies. It’s currently testing a smaller version of Gemini with them.

Will Gemini cost money to use?

At this point, we don’t know if Google will charge to use Gemini. It seems possible, however. Bard is free to use, so it would make sense for Google to charge for the more powerful model. We’ll need to wait for the official launch to know for sure.

Will Gemini be better at preventing hallucinations?

This is an important area for AI. Hallucinations occur when an AI model generates facts out of thin air. These facts are not based on any actual information, and it’s almost always completely wrong. This is what happened when Bard was unveiled.

According to a person who was able to try out an early version of Gemini, it seems to be better for avoiding hallucinations.

Will Gemini replace Google Bard?

Not likely. Bard has already been implemented into several Google products. It could be similar to what OpenAI is doing. The company has ChatGPT 3.5 as the lightweight consumer-facing product and GPT-4 as the more advanced one. OpenAI is still developing both.

This could be the case with Gemini. Bard could be the user-facing product that you’ll use in your daily life while Gemini could be the more advanced product to be used for more advanced tasks.

How many versions will there be?

At the moment, we don’t know how many versions of Gemini Google will launch. We know that it will launch a smaller version of the model to the general public. This could be for the average user. However, the company is rumored to launch a much larger version for more professional use.

When will it launch?

[Updated, December 4th: Gemini is expected to launch in January 2024]

Currently, we don’t have an exact date for when Google is going to release Gemini. The company is currently testing a version of it with different companies. This could point to a release in the near future. We know that the company plans to launch it sometime later this year.


[ad_2]
Source link

Social media giants to testify over failing to protect kids

0
[ad_1]

US senators have urgently invited the CEOs of five of the major social media giants to testify about their failure to protect children online. The Senate Judiciary Committee said it will hear from Meta CEO Mark Zuckerberg, X (formerly Twitter) CEO Linda Yaccarino, TikTok CEO Shou Zi Chew, Snap CEO Evan Spiegel, and Discord CEO Jason Citron.

In a press release, the US senate committee on the judiciary announced that the Committee’s previously announced hearing on online child sexual exploitation has been rescheduled for January 31, 2024 and will feature testimony from the CEOs. An earlier hearing on Tuesday, February 14, 2023, included only consumer advocates as witnesses, and no industry representatives.

The CEOs of X, Discord, and Snap will testify after subpoenas were issued by the Committee, following repeated refusals by the three leaders to testify. The CEOs of Meta and TikTok voluntarily agreed to testify at the hearing.

Senators Durbin and Graham commented:

“Several companies initially refused to accept a subpoena. The US Marshals Service even attempted to serve the subpoena at Discord’s office. Both actions are remarkable departures from typical practice.”

The hearing comes as part of a bipartisan effort to protect children online. To that end, several online safety bills across multiple states have gone into effect. For example, Utah signed a bill in March that will require minors to obtain parental consent to sign up to social platforms, while both Louisiana and Mississippi now require age verification to view content considered harmful to children, like porn. On the other hand, a federal judge has blocked a Texas law requiring age verification and a health warning for viewing pornographic websites, a day before the law was set to take effect.

In May, we talked to Alec Muffet about the possible downsides of some of these bills.

During the “Protecting our children online” hearing in February, witnesses and senators mentioned requirements like parental controls, default settings, and audits as tools that could be used to promote online safety for teenagers.  They focused on the importance of holding platforms liable for failure to enforce their own terms, and discussed imposing a duty of care on online platforms.

So now seems to be the time that the CEOs of the major platforms will be forced to explain what they have done in the past and how they plan to do better in the future. You would expect they would like to bring their ideas and input voluntarily to the table, but nonetheless it took subpoenas to get them all there.

Children and online safety

The internet is both a good and bad place. A good approach is to spend little to no time on sites that do not give your child a positive and learning experience. And when it comes to internet safety for kids and teens, the best approach is for parents and carers to be involved in their child’s digital life.

If you don’t want to rely on the introduction of legislation and how the social media platforms will undoubtedly struggle to become compliant, we can recommend reading our blog titled “Internet safety tips for kids and teens: A comprehensive guide for the modern parent.”

I do expect some of the platforms to drag their feet, because it seems they always do. Meta is already facing a lawsuit, filed in a California federal court, which argues that Meta unlawfully misled the public about the harms its products, like Facebook and Instagram, could impose on children and teens.

In the UK, Bytedance’s TikTok is looking at a $28.91m fine related to how children are safeguarded on the app.

And Meta, ByteDance, Alphabet, and Snap, are facing another lawsuit alleging their social platforms have adverse mental health effects on children and for running platforms that are addictive to kids.

While it is clear that something needs to be done to protect our children, agreeing on the way in which we can achieve this is hard. Especially if we can’t rely on all the social media platforms to volunteer their cooperation.


We don’t just report on threats—we remove them

Cybersecurity risks should never spread beyond a headline. Keep threats off your devices by downloading Malwarebytes today.


[ad_2]
Source link

Google is ditching ‘Messages by Google’ name for something simpler

0
[ad_1]

Google is ditching its ‘Messages by Google’ name for something simpler, and it was about time. It’s not difficult to guess what is going on here, Google will now use the ‘Google Messages’ name, finally.

Google is finally ditching ‘Messages by Google’ name

The company has been using the ‘Messages by Google’ name for a long time now, but it simply doesn’t sound right as an app name. Pretty much everyone referred to it as Google Messages, and Google is finally embracing it.

Google recently rolled out seven new features to the app, and along with that, changed the moniker too… quietly. The change occurred in the Google Play Store, as the app listing simply says ‘Google Messages’ now.

Google did use this “new” name in its blog post too, the one announcing new features for the app. This change is actually nice to see, and many people will likely agree.

One thing to note is that the app is still called ‘Messages’ on Android, without the ‘Google’ part. Let’s hope Google will make a change to its ‘Phone’ app soon too. That one is still referred to as ‘Phone by Google’ instead of ‘Google Phone’, or something like that.

Google Messages is the go-to SMS & RCS app for many users

Google Messages is basically a default SMS and RCS app on Android. It comes pre-installed on Google’s very own phones, and a number of other Android smartphones. Those that do not have it, can easily grab it from the Google Play Store.

Considering that Apple will roll out RCS support next year, this app will almost certainly become even more prominent. Google is constantly adding new features, and the app keeps getting better and better as a result.

The app is, of course, free to use, and those of you who still don’t have it installed on your phone can grab it from the Google Play Store.


[ad_2]
Source link

Spotify makes significant workforce reduction, laying off 17% of its employees

0
[ad_1]

As the end of the year approaches, Spotify managed to deliver joy to its more than 550 million users with Spotify Wrapped and its revamped design and features. But by the end of 2023, not all the people working at the Swedish company might be in a cheerful mood.

In a surprise move, Spotify announced a significant reduction in its workforce, shedding 17% of its employees. The decision comes as the music-streaming giant grapples with economic uncertainties and seeks to streamline its operations.

In a company press release (via Engadget), CEO Daniel Ek acknowledged the challenging times ahead and explained the rationale behind the layoffs. He stated that the cuts were necessary to align Spotify’s cost structure with its long-term goals and ensure sustainable growth.

Despite recent positive earnings reports, Ek emphasized the need for decisive action to address the widening gap between Spotify’s financial targets and its current operational costs. He acknowledged the impact on employees, expressing empathy for those affected. “While I am convinced this is the right action for our company, I also understand it will be incredibly painful for our team,” he said.

The layoffs follow a period of rapid expansion for Spotify, driven by favorable investment conditions in 2020 and 2021. While these investments contributed to Spotify’s growth, Ek acknowledged that the company’s cost structure remains too high.

Approximately 1,500 employees are expected to be affected by the layoffs, reducing Spotify’s workforce from around 9,000 employees. To mitigate the impact, Spotify offers a five-month severance package, healthcare coverage during that period, and immigration/career support.

Ek emphasized the importance of lean operations in Spotify’s next phase, stating that it is not just an option but a necessity. He also highlighted the company’s revamped royalty model, designed to better support artists while reducing fraudulent streams.

Spotify’s decision reflects the broader challenges facing the tech industry, with many companies facing economic headwinds and being forced to make difficult decisions. Nokia’s recent announcement to slash 14,000 jobs and tech giants like Google, Amazon, Microsoft, and Meta laying off over 50,000 employees collectively in 2023 underscore the industry’s struggles.


[ad_2]
Source link

OnePlus Open survives a rigorous durability test: Video

0
[ad_1]

The OnePlus Open, a $1,699 foldable device was durability tested by Zack Nelson from JerryRigEverything. For context, or OnePlus’s track records in his tests, the OnePlus 11‘s back glass was shattered and the OnePlus 10 Pro was cracked in half. The OnePlus Open is subject to a million folds, a feat proclaimed by the company and demonstrated in a controlled environment.

The durability testing of the OnePlus Open begins with a scrutiny of the surface or the external elements.

In the scratch test, the Ceramic Guard glass (used on the cover screen) only suffers from slight scratches when scratched with a level 7 element on the Mohs scale of hardness. This is not a big deal, and it keeps parity with most other flagships in the market. Not to mention, with the aluminum side frame and the glass front and back, the phone appears to be sturdy while closed.

The soft inner screen can suffer from visible marks just from nails: be aware

However, for a foldable, the story starts when you unfold the foldable. Opening up the phone exposes the soft plastic display. Pressing a little hard on the inner display with nails leads to visible marks on it.

Notably, OnePlus says the phone has a layer of ultra-thin glass on the inner display panel. However, the whole display assembly includes additional layers such as the one for touch response. On top of it, is a soft plastic layer, always, for every foldable launched to this date.

Coming to ingress protection, the OnePlus Open is IPX4, meaning it’s not officially dust-resistant. The ‘4’ rating for water splash protection, is inferior in comparison to Samsung’s Galaxy Z Fold 5 or Google Pixel Fold’s ‘8’ or submersion rating. Speaking of dust, the phone does not have any official resistance rating against it.

The reviewer exposes the hinge to a plethora of dust and even larger granular particles. But surprisingly, the hinge appears to work just fine after the torture.

Opening up the OnePlus Open unveils the Open’s resilience against Nelson’s capabilities

After surviving the scratches and dust, the reviewer bends the phone (to the maximum of his capabilities) in both the closed and opened states to ensure that you won’t accidentally damage the phone while keeping it between you and the sofa.

When being folded in the wrong direction, the phone bends a bit but locks out and doesn’t snap in half. The OnePlus Open survives the bent test. It remains fully functional and passes JerryRigEveruthing’s extreme durability test.


[ad_2]
Source link

WhatsApp to integrate with Instagram, allowing users to share status updates

0
[ad_1]
WhatsApp, the popular messaging app with over 2.5 billion active users, is set to enhance its connection with Instagram by introducing a new feature that allows users to seamlessly share status updates across both platforms. Building upon its previous integration with Facebook, it looks like WhatsApp is now extending this functionality to Instagram, further streamlining the process of sharing content with a wider audience.

According to WABetaInfo, this upcoming feature, which is currently under development, will enable users to effortlessly share their WhatsApp status updates directly to their Instagram stories. By eliminating the need to manually create and post separate updates on each platform, this integration will improve content sharing.


The time-saving benefits of this feature are particularly noteworthy. Instead of dedicating time to creating and posting identical updates on both WhatsApp and Instagram, users will be able to do this task in a single step, streamlining their social media engagement.


Beyond convenience, this integration also promotes consistency in content sharing across Meta’s platforms. Users can ensure that their updates align with their overall social media presence, maintaining a cohesive identity for their audience. Moreover, the optional nature of this feature empowers users to decide whether or not they want to use it. Those who prefer to maintain separate status updates on each platform can do so without restrictions.


The feature to share status updates to Instagram is found in the latest WhatsApp beta for Android 2.23.25.20 update, which is available on the Google Play Store. It is still under development, and it will be available in a future update of the app.


[ad_2]
Source link

A week in security (November 27 – December 3)

0
[ad_1]

December 1, 2023 – Domain fronting is a technique to hide the true origin of HTTPS requests by hiding the real domain name encrypted inside a legitimate TLS request.

November 30, 2023 – A fake antivirus alert may suddenly hijack your screen while browsing. This latest malvertising campaign hit top publishers.

November 29, 2023 – Google’s released an update to Chrome which includes seven security fixes. Make sure you’re using the latest version!

November 28, 2023 – A vulnerability in the ownCloud file sharing app could lead to the exposure of sensitive credentials like admin passwords.

November 23, 2023 – Google has set a date for the introduction of Manifest V3 which will hurt the capabilities of many ad blockers.


[ad_2]
Source link

OnePlus 12 appears in hands-on images with 24GB of RAM

0
[ad_1]

OnePlus’ new flagship is right around the corner. The OnePlus 12 will launch tomorrow, and its hands-on images have just appeared online. Some of the images below come from OnePlus China President himself, while others are leaks, you’ll spot the difference easily. Even a very short hands-on video popped up for everyone to see.

The OnePlus 12 hands-on images give us an even better look at the phone

The phone’s design is far from being a secret, as OnePlus itself showed us the phone, in detail. We’ve seen a ton of renders and lifestyle images already. The images that just leaked only confirm what we already knew, pretty much.

If you check out the gallery below, you’ll see real-life images of the device, in both white and green colors. Both of these phones seem to have glass on the back, though we can’t confirm that.

Official real-life images:

Leaked real-life images:

The white variant has a glossy backplate, while the green one does not

The white variant is glossy on the back, while the green one doesn’t seem to be. The finish on the back actually looks like marble, kind of. A circular camera setup sits on the back, and it’s connected to the side with a piece of metal, for design purposes, we presume. There are three cameras included on the back.

The phone’s display is curved, as you can see in the provided images. A display camera hole is centered this time around, while the bezels are extremely thin. It’s even possible that they’re uniform, we’ll have to wait and see.

A 5,400mAh battery is confirmed, while this is a 24GB RAM model

In one of the images, you’ll even see that the battery capacity has been confirmed thanks to the About Phone screen. A 5,400mAh battery is coming, and this is a 1TB version of the phone. A 6.82-inch display is mentioned, and the phone’s cameras too. 50-megapixel, 64-megapixel, and 48-megapixel rear-facing units.

This particular phone also has 24GB of RAM, and you can even add 4GB more via virtual RAM. This variant will most likely be sold in China only, though.

The OnePlus 12 is launching tomorrow in China, while the global launch event will take place in January. January 23 is the rumored date.


[ad_2]
Source link

Man lost $280,000 because of ‘FjordPhantom’ Android virus

0
[ad_1]

Cybercrime has been on the rise in recent years, and Android users have become a prime target for attackers. This is due in part to the open nature of the Android platform. It gives users more control over their devices but also makes them more vulnerable to malware and other threats. In a recent example, a Thai victim lost around 10 million Thai Baht (roughly $280,000) to cybercriminals deploying FjordPhantom. More importantly, it is roughly 4 times the annual salary of an average American, per the 2022 census.

The Android trojan draining bank accounts is FjordPhantom

FjordPhantom malware was First discovered by security firm Promon. That malware is wreaking havoc across several Asian countries, including Indonesia, Thailand, Vietnam, Singapore, and Malaysia. The attack starts when someone sends a message or email to a person, asking them to download a real banking app.

However, there’s a trick — once the app is downloaded, it runs in a special virtual environment that lets the attackers take control. This virtualization thing helps them make the same app work for two people on one device.

The attackers use a special virtualization technique to make the fake app work just like the real app. This means the fake app can run on the same phone as the real one without any problems. They also use hooking to change how the phone’s system works. This allows them to control what happens when you use certain apps. Finally, the attackers call people who have downloaded the fake app and pretend to be from the bank. They try to trick people into giving them their passwords or making transactions. That results in people losing money, of course.

How to be on the safer side against such cybercrimes

There’s no foolproof advice to simply avoid cybercrime; however, there are a few steps to be on the safer side. For example, keeping the phone’s software updated. Google frequently pushes security updates to Android phones. Another move is to refrain from opening suspicious attachments or clicking on unknown links. Whenever possible, enable two-factor authentication for your online accounts. One more advice is to rely on Google Play Store or other leading app stores for installing apps.


[ad_2]
Source link

YouTube will now show you the number of views and likes a video has in real-time

0
[ad_1]
Google is changing the “likes” and “views” counters on YouTube so that they will refresh automatically when a video is viewed. Currently, if you’re watching a video on the platform, the number of likes and views won’t change while you are viewing the clip. But with this update, those two metrics will rise and show a cool animation of a rolling numbers counter whether the video is being played or is paused. And the animation will be played even if you switch to another app and return to YouTube.
Android Police came across this new behavior on YouTube and couldn’t find any reference to it except for a post on Reddit from back in October from user nonotan who wrote, “I’m watching a video, and I keep getting distracted by an animation where the view counter is rolling over to a new number as if it were a physical click counter. I have not seen this behavior before today, and I’d like to disable it.”

The same Redditor was so bothered by the effect that he added the following line to his filters which stopped the animation and the count from advancing in real-time: https://www.youtube.com/youtubei/v1/updated_metadata*.

Others also had similar posts about the feature such as Reddit user Reyh who said, “Distracting indeed and imho not beneficial at all. Saw it on Tuesday for the first time, using Firefox. The first few times I only noticed some sort of movement at the bottom left of the videos. I wondered if it was in the video itself. Then I observed the area for a while and saw the new counter animation.”

It seems that this feature is being tested with the help of a very limited number of YouTube users globally. Those who see the feature will notice the number of views and the number of likes rolling up on the screen to a higher number while they are viewing the video. Google calls this progress and for some of you, it may be.

[ad_2]
Source link